Optical semiconductor device and method for manufacturing optical semiconductor device

Abstract

The invention provides an optical semiconductor device and a manufacturing method of the optical semiconductor device. The optical semiconductor device can be installed with high precision by suppressing position deviation. An optical semiconductor device according to one embodiment includes: a first semiconductor element having a first bonding surface and an end surface that intersects the first bonding surface and is capable of emitting an optical signal; and a second semiconductor element having a second bonding surface facing the first bonding surface and an optical waveguide extending in a direction parallel to the second bonding surface and capable of transmitting an optical signal. The first bonding surface and the second bonding surface are hydrophilically bonded to each other, and the end surface of the first semiconductor element and the optical waveguide of the second semiconductor element are optically coupled to each other.

Description

Technical Field

[0001] The present invention relates to an opto-semiconductor device and a method for manufacturing the opto-semiconductor device. Background Art

[0002] Patent Document 1 describes a semiconductor light-emitting element and a method for manufacturing the same. The semiconductor light-emitting element includes: a first n-side nitride semiconductor layer; a first active layer provided on the first n-side nitride semiconductor layer; a first light-emitting portion provided on the first active layer and including at least a first p-side nitride; and a second n-side nitride semiconductor layer directly bonded to the first light-emitting portion. The first light-emitting portion has a first bonding surface, and the second n-side nitride semiconductor layer has a second bonding surface. The first bonding surface is directly bonded to the second bonding surface. The first bonding surface is in direct contact with the second bonding surface without using a resin or an adhesive. As the direct contact, cases of bonding by direct contact such as a surface activation bonding method and an atomic diffusion bonding method are described.

[0003] Patent Document 2 describes an optical module. The optical module is an integrated laser module in which an LD element, a driver IC, an optical fiber, and a sub-substrate are mounted on the upper surface of a silicon substrate, i.e., an Si platform. In the LD element, the driver IC is mounted by soldering. The LD element on which the driver IC is mounted is mounted on the upper surface of the Si platform by surface activation bonding. The Si platform has a bonding portion for bonding the sub-substrate and the LD element. The sub-substrate and the LD element are respectively surface activation bonded to the bonding portion of the Si platform.

[0004] Patent Document 3 describes an optical module in which a wavelength conversion element as an optical element is mounted on a silicon substrate. The silicon substrate has a bonding portion in the form of a micro bump structure made of gold. The wavelength conversion element is bonded to the bonding portion by a surface activation bonding technique.

[0005] Patent Document 4 describes a laser chip having a flip-chip mounted light source on a silicon photonics chip having an optical waveguide. The silicon photonics chip and the laser chip each have a vertical stopper. Alignment of the laser chip with respect to the silicon photonics chip is performed by docking the two vertical stoppers with each other vertically. The laser chip is mounted on the silicon photonics chip by soldering.

[0006] Prior Art Documents Patent Documents Patent Document 1: Japanese Unexamined Patent Application Publication No. 2022-132130 Patent Document 2: International Publication No. 2015 / 146377 Patent Document 3: Japanese Unexamined Patent Application Publication No. 2013-228691 Patent Document 4: U.S. Patent Application Publication No. 2022 / 0123518 Summary of the Invention Problems to be Solved by the Invention In the case where a laser chip is mounted on a silicon photonics chip by solder as described above, after aligning the laser chip with respect to the silicon photonics chip, the solder is heated. When this heating is performed, sometimes a positional shift of the laser chip occurs. If a positional shift of the laser chip occurs, the efficiency of optical coupling may decrease. Therefore, it is required to suppress the positional shift and perform the mounting with high precision.

[0007] An object of the present invention is to provide an optoelectronic device and a method for manufacturing the optoelectronic device that can suppress a positional shift and perform the mounting with high precision.

[0008] Means for Solving the Problems The optoelectronic device according to the present invention includes: a first semiconductor element having a first bonding surface and an end surface that intersects the first bonding surface and can emit an optical signal; and a second semiconductor element having a second bonding surface facing the first bonding surface and an optical waveguide that extends in a direction parallel to the second bonding surface and can transmit an optical signal. The first bonding surface and the second bonding surface are hydrophilically bonded to each other, and the end surface of the first semiconductor element and the optical waveguide of the second semiconductor element are optically coupled to each other.

[0009] Advantages of the Invention According to the present invention, it is possible to suppress a positional shift and perform the mounting with high precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a perspective view showing an optoelectronic device according to a first embodiment.

[0011] Figure 2 is a cross-sectional view of the optoelectronic device cut by a plane orthogonal to the optical waveguide of the second semiconductor element of the optoelectronic device of Figure 1 .

[0012] Figure 3 schematically shows Figure 2 a cross-section of the first semiconductor element in the optoelectronic device of

[0013] Figure 4 is a view showing Figure 3 a first modification of the cross-section of the first semiconductor element of

[0014] Figure 5 is a view showing Figure 3 a second modification of the cross-section of the first semiconductor element of

[0015] Figure 6 is a view showing Figure 3 a third modification of the cross-section of the first semiconductor element of

[0016] Figure 7 It is a diagram showing Figure 3 a fourth modification of the cross-section of the first semiconductor element.

[0017] Figure 8 It is a diagram showing Figure 3 a fifth modification of the cross-section of the first semiconductor element.

[0018] Figure 9 It is a diagram for explaining the hydrophilic treatment for the first semiconductor element and the second semiconductor element.

[0019] Figure 10 It is a diagram for explaining the hydrophilic treatment, temporary bonding, and formal bonding in the manufacturing method of the opto-semiconductor device according to the first embodiment.

[0020] Figure 11 It is a diagram showing the process of the manufacturing method of the opto-semiconductor device according to the first embodiment.

[0021] Figure 12 It is a perspective view of the opto-semiconductor device according to the second embodiment.

[0022] Figure 13 It is a cross-sectional view of the opto-semiconductor device cut by a plane orthogonal to the optical waveguide of the second semiconductor element of the Figure 12 opto-semiconductor device.

[0023] Figure 14 It schematically shows Figure 13 the cross-section of the first semiconductor element in the

[0024] Figure 15 opto-semiconductor device. Figure 14 It is a diagram showing the cross-section of a modification of the first semiconductor element.

[0025] Figure 16 It is a diagram showing Figure 15 a modification of the cross-section of the first semiconductor element.

[0026] Figure 17 It is a diagram showing Figure 15 a modification of a different cross-section of the first semiconductor element of Figure 16 the first semiconductor element.

[0027] Figure 18 It is a diagram showing the process of the manufacturing method of the opto-semiconductor device according to the second embodiment.

[0028] Figure 19 It is a diagram showing the process of the manufacturing method of the opto-semiconductor device according to the third embodiment.

[0029] Figure 20 It is a diagram showing the optoelectronic semiconductor device according to the third embodiment.

[0030] Figure 21 It is a diagram showing the manufacturing process of the optoelectronic semiconductor device according to the fourth embodiment.

[0031] Figure 22 It is a diagram showing the manufacturing process of the optoelectronic semiconductor device according to the fifth embodiment.

[0032] Figure 23 It is a perspective view showing the optoelectronic semiconductor device according to the sixth embodiment.

[0033] Figure 24 It is Figure 23 a perspective view showing an enlarged view of the first bonding agent layer and the second bonding agent layer of the optoelectronic semiconductor device.

[0034] Figure 25 It is showing Figure 23 a perspective view of the second semiconductor element of the optoelectronic semiconductor device.

[0035] Figure 26 It is Figure 25 a cross-sectional view showing an enlarged view of a part of the second semiconductor element.

[0036] Figure 27 It is Figure 25 a cross-sectional view showing an enlarged view of a part of the second semiconductor element.

[0037] Figure 28 It is showing Figure 23 cross-sectional views of the optoelectronic semiconductor device before and after solder melting.

[0038] Figure 29 It is showing Figure 23 cross-sectional views of the optoelectronic semiconductor device before and after solder melting.

[0039] Figure 30 It is a diagram showing the manufacturing process of the optoelectronic semiconductor device according to the sixth embodiment.

[0040] Figure 31 It is a perspective view showing the optoelectronic semiconductor device according to the seventh embodiment.

[0041] Figure 32 It is showing Figure 31 a perspective view of the second optoelectronic semiconductor element of the optoelectronic semiconductor device.

[0042] Figure 33 It is Figure 32 a cross-sectional view of the second semiconductor element.

[0043] Figure 34 It is a cross-sectional view showing an enlarged part of an Figure 31 opto-semiconductor device.

[0044] Figure 35 It is a cross-sectional view showing an Figure 31 opto-semiconductor device before and after solder melting.

[0045] Description of reference numerals 1, 1A, 1B, 1C, 1D, 1E: optical semiconductor device; 10, 10A, 10B, 10C: first semiconductor element; 11: end face; 12: electrode; 13: n-InP layer; 13b: convex portion; 13d: surface; 14, 14B: p-InP layer; 14b: convex portion; 14c: convex portion; 14d: surface; 14f: surface; 15: MQW layer; 16: p-InP layer; 17: electrode; 17b: extension portion; 18, 18A, 18B, 18C: first bonding surface; 20: second semiconductor element; 21: first layer; 22: second layer; 23: third layer; 24: optical waveguide; 25: electrode; 26: second bonding surface; 27: support portion; 28: electrode; 29: recessed portion; 30, 30A: joint portion; 31, 31A: oxide film; 32: electrical bonding agent; 40, 40A: first semiconductor element; 44: p-InP layer; 44b: concave portion; 44c: first side portion; 44d: bottom; 44f: second side portion; 45: bonding layer; 46: p-InP layer; 47: n-InP layer; 48, 48A: first bonding surface; 50: second semiconductor element; 56: second bonding surface; 57: support portion; 60: first semiconductor element; 70: second semiconductor element; 80: first semiconductor element; 90: second semiconductor element; 110: first semiconductor element; 120: second semiconductor element; 125: electrode; 126: first bonding agent layer; 126A: first bonding agent layer region; 126b: first part; 126c: second part; 126d: third part; 127: second bonding agent layer; 127A: second bonding agent layer region; 127b: first part; 127c: second part; 127d: third part; 130: bonding agent layer; 131: third bonding surface; 132: fourth bonding surface; 140: second semiconductor element; 145: electrode; 145b: first electrode; 145c: second electrode; 145d: first part; 145f: second part; 145h: first part; 145j: second part; 146: second bonding surface; 146b: first small bonding surface; 146c: second small bonding surface; 147: support portion; 149A: first bonding agent layer region; 150A: second bonding agent layer region; 160: bonding agent layer; 161: third bonding surface; 162: fourth bonding surface; D1: first direction; D2: second direction; D3: third direction (cross direction); L: distance; S: gap; W1: bonding wire. Detailed implementation manners

[0046] [Description of the embodiments of the invention of the present application] First, the contents of the embodiments of the present invention will be described. (1) The optoelectronic semiconductor device according to one embodiment includes: a first semiconductor element having a first bonding surface and an end surface that intersects the first bonding surface and is capable of emitting an optical signal; and a second semiconductor element having a second bonding surface opposed to the first bonding surface and an optical waveguide that extends in a direction parallel to the second bonding surface and is capable of transmitting an optical signal. The first bonding surface and the second bonding surface are hydrophilically bonded to each other, and the end surface of the first semiconductor element and the optical waveguide of the second semiconductor element are optically coupled to each other.

[0047] (1) The manufacturing method of the optoelectronic semiconductor device according to one embodiment includes the following steps: hydrophilizing the first bonding surface of the first semiconductor element; performing alignment so that the second bonding surface of the second semiconductor element is opposed to the first bonding surface in a separated state, and optically coupling the end surface of the first semiconductor element capable of emitting an optical signal and the optical waveguide of the second semiconductor element capable of transmitting an optical signal to each other; bringing the first bonding surface and the second bonding surface into contact at a first temperature and pressing at least one of the first semiconductor element and the second semiconductor element to temporarily bond them to each other; and after the step of temporarily bonding, heating the first semiconductor element and the second semiconductor element to formally bond the first bonding surface and the second bonding surface to each other at a second temperature higher than the first temperature.

[0048] In this optoelectronic semiconductor device and the manufacturing method of the optoelectronic semiconductor device, the first semiconductor element has a first bonding surface and an end surface, and the second semiconductor element has a second bonding surface and an optical waveguide. The end surface of the first semiconductor element and the optical waveguide of the second semiconductor element are optically coupled to each other, and the first bonding surface and the second bonding surface are hydrophilically bonded to each other. In the hydrophilic bonding, heating after alignment of the first bonding surface with respect to the second bonding surface is not required. Therefore, the position of the first semiconductor element relative to the second semiconductor element can be suppressed from shifting due to heating. Accordingly, the mounting of the first semiconductor element relative to the second semiconductor element can be performed with high precision while suppressing the position shift.

[0049] (2) Alternatively, on the basis of the above (1), an oxide film may be formed at the joint portion which is the portion where the first bonding surface and the second bonding surface are hydrophilically bonded to each other. In this case, the first bonding surface and the second bonding surface can be firmly bonded by the oxygen atoms constituting the oxide film.

[0050] (3) Alternatively, based on (1) or (2) above, the first semiconductor element has a third joint surface parallel to the first joint surface, and the second semiconductor element has a fourth joint surface facing the third joint surface. The third joint surface and the fourth joint surface are joined to each other by an adhesive at the joint portion. The joint portion is formed in a region where the third joint surface and the fourth joint surface overlap when viewed from the direction crossing the third joint surface, i.e., the crossing direction. Alternatively, the opto-semiconductor device further includes an adhesive layer formed on at least one of the third joint surface and the fourth joint surface. The adhesive layer is composed of an adhesive. At the joint portion, the adhesive layer is filled between the third joint surface and the fourth joint surface. Alternatively, the area of the adhesive layer when viewed from the crossing direction is larger than the area of the joint portion when viewed from the crossing direction. Based on the state where the adhesive layer is filled between the third joint surface and the fourth joint surface at the joint portion, the area of the adhesive layer is larger than the area of the joint portion, whereby the third joint layer and the fourth joint layer can be reliably joined to each other at the joint portion. The area of the adhesive layer is, for example, equal to or less than the area of the fourth joint layer.

[0051] (4) Alternatively, in (1) or (2) above, the second joint surface has a first small joint surface and a second small joint surface that are separated from each other along a direction parallel to the second joint surface, i.e., the first direction. Alternatively, the first semiconductor element has a third joint surface between the first small joint surface and the second small joint surface, the third joint surface being parallel to the first joint surface and extending in a direction crossing the first direction, i.e., the second direction, and the second semiconductor element has a fourth joint surface facing the third joint surface and extending in the first direction. The third joint surface and the fourth joint surface are joined to each other by an adhesive at the joint portion. The joint portion is formed in a region where the third joint surface and the fourth joint surface overlap when viewed from the crossing direction. Alternatively, the opto-semiconductor device further includes an adhesive layer formed on at least one of the third joint surface and the fourth joint surface. The adhesive layer is composed of an adhesive. At the joint portion, the adhesive layer is filled between the third joint surface and the fourth joint surface. Alternatively, the area of the adhesive layer when viewed from the direction crossing the third joint surface, i.e., the crossing direction, is larger than the area of the joint portion when viewed from the crossing direction. Based on the state where the adhesive layer is filled between the third joint surface and the fourth joint surface at the joint portion, the area of the adhesive layer is larger than the area of the joint portion, whereby the third joint layer and the fourth joint layer can be reliably joined to each other at the joint portion. It should be noted that when the adhesive layer is formed on the third joint layer, the area of the adhesive layer is equal to or less than the area of the third joint layer, and when the adhesive layer is formed on the fourth joint layer, the area of the adhesive layer is equal to or less than the area of the fourth joint layer.

[0052] (5) A method for manufacturing an opto-semiconductor device according to an embodiment includes the following steps: hydrophilizing a first bonding surface of a first semiconductor element; performing alignment so that a second bonding surface of a second semiconductor element faces the first bonding surface in a separated state, and optically coupling an end face of the first semiconductor element capable of emitting an optical signal and an optical waveguide of the second semiconductor element capable of transmitting the optical signal; bringing the first bonding surface into contact with the second bonding surface at a first temperature and pressing at least one of the first semiconductor element and the second semiconductor element to temporarily bond them to each other; and after the step of performing the temporary bonding, heating the first semiconductor element and the second semiconductor element to perform a formal bonding between the first bonding surface and the second bonding surface at a second temperature higher than the first temperature.

[0053] (6) Alternatively, based on the above (5), in the step of hydrophilizing, the first bonding surface is hydrophilized by irradiating ultraviolet rays on the first bonding surface in an air-exposed environment. In this case, it is possible to suppress damage to the first bonding surface when hydrophilizing the first bonding surface. Thereby, it is possible to reduce the deterioration of the hydrophilic bonding due to the roughening of the first bonding surface.

[0054] (7) Alternatively, based on the above (5), in the step of hydrophilizing, the first bonding surface is hydrophilized by exposing the first bonding surface to oxygen plasma in a vacuum environment. In this case, when hydrophilizing the first bonding surface, a thick oxide film can be formed on the first bonding surface. The thick oxide film can enhance the bonding force.

[0055] (8) Alternatively, in the above (5), in the step of hydrophilizing, the first bonding surface is hydrophilized by exposing the first bonding surface to nitrogen plasma in a vacuum environment.

[0056] (9) Alternatively, based on any one of the above (5) to (8), the strength of a joint portion, which is a portion where the first bonding surface and the second bonding surface are joined to each other after the step of performing the temporary bonding, is 5 MPa or more. In this case, the first semiconductor element can be firmly bonded to the second semiconductor element during the temporary bonding.

[0057] (10) Alternatively, based on any one of the above (5) to (9), the distance between the first bonding surface and the second bonding surface during the alignment step is 1 μm or more and less than 100 μm. In this case, it is possible to reduce the distance between the first bonding surface and the second bonding surface during alignment, and thus it is possible to more reliably suppress the positional deviation of the first bonding surface with respect to the second bonding surface.

[0058] (11) Alternatively, based on any one of (5) to (10) above, the first temperature is 20°C or higher and 40°C or lower, and the second temperature is 100°C or higher and 300°C or lower. In this case, since the first temperature during temporary bonding can be set to room temperature, it is possible to more reliably suppress the positional deviation due to heating.

[0059] (12) Alternatively, based on any one of (5) to (11) above, the first semiconductor element has a third bonding surface parallel to the first bonding surface, and the second semiconductor element has a fourth bonding surface facing the third bonding surface. Alternatively, the manufacturing method may include a step of forming a first bonding agent layer made of an electrical bonding agent on the fourth bonding surface and a step of forming a second bonding agent layer made of an electrical bonding agent on the first bonding agent layer before the alignment step. Alternatively, the manufacturing method may include a step of melting the first bonding agent layer and the second bonding agent layer by heating, and bonding the third bonding surface and the fourth bonding surface to each other at the bonding portion by the melted first bonding agent layer and the second bonding agent layer, i.e., the bonding agent layer. In this case, by filling the melted first bonding agent layer and the second bonding agent layer, i.e., the bonding agent layer, between the third bonding layer and the fourth bonding layer at the bonding portion, the third bonding surface and the fourth bonding surface can be reliably bonded to each other at the bonding portion.

[0060] [Details of Embodiments of the Present Invention] Hereinafter, specific examples of the optical semiconductor device and the manufacturing method of the optical semiconductor device according to the embodiments of the present invention will be described with reference to the drawings. It should be noted that the present invention is not limited to the following examples, and is intended to include all modifications within the scope equivalent to the claims shown in the claims. In the description of the drawings, the same or corresponding elements are denoted by the same reference numerals, and repeated descriptions are appropriately omitted. For ease of understanding, the drawings sometimes depict a part in a simplified or exaggerated manner, and the dimensional ratios, etc. are not limited to those described in the drawings.

[0061] Figure 1 is a perspective view showing the optical semiconductor device 1 according to the present embodiment. As Figure 1As shown, the opto-semiconductor device 1 has a rectangular parallelepiped shape. The opto-semiconductor device 1 includes a first semiconductor element 10 and a second semiconductor element 20. For example, the second semiconductor element 20 is a silicon photonics chip, and the first semiconductor element 10 is an opto-semiconductor element fixed to the second semiconductor element 20. The opto-semiconductor element is, for example, a semiconductor laser element or a semiconductor optical amplifier element. The semiconductor laser element and the semiconductor optical amplifier element supply light to the silicon photonics chip as a light source. Hereinafter, the direction in which the first semiconductor element 10 is provided when viewed from the second semiconductor element 20 may be referred to as up, the upper side, or the upper direction, and the opposite direction may be referred to as down, the lower side, or the lower direction. However, these directions are for convenience of explanation and do not limit the arrangement position or direction of the object, etc.

[0062] For example, the second semiconductor element 20 has a rectangular parallelepiped shape. Figure 2 It is a cross-sectional view of the first semiconductor element 10 and the second semiconductor element 20 cut by a plane extending in both the up-down direction and the width direction of the second semiconductor element 20. As Figure 1 and Figure 2 shown, the first semiconductor element 10 has an end face 11 capable of emitting an optical signal. The second semiconductor element 20 has an optical waveguide 24 capable of transmitting an optical signal. The end face 11 and the optical waveguide 24 are optically coupled to each other.

[0063] For example, the second semiconductor element 20 has a first layer 21, a second layer 22 located above the first layer 21, and a third layer 23 located above the second layer 22. The first layer 21 is, for example, a silicon (Si) layer. The first layer 21 has a rectangular parallelepiped shape. The second semiconductor element 20, for example, has a plurality of second layers 22 and a plurality of third layers 23. The second layer 22 extends along the first direction D1, which is the direction in which the optical waveguide 24 extends, and the second direction D2 described later. In addition, the third layer 23 extends along the first direction D1 and the second direction D2. When viewed from the end face 11, along the first direction D1, which is the direction in which the optical waveguide 24 is provided, the two second layers 22 are separated from each other. The two third layers 23 are separated from each other along the first direction D1. Here, an optical waveguide 24 is formed in one of the third layers 23, and no optical waveguide 24 is formed in the other third layer 23. One of the second layers 22 is located below one of the third layers 23, and the other second layer 22 is located below the other third layer 23. It should be noted that the two second layers 22 separated along the first direction D1 may also be formed integrally with each other. The two third layers 23 separated along the first direction D1 may also be formed integrally with each other.

[0064] Two of the plurality of second layers 22 in which no optical waveguides 24 are formed are separated from each other in a second direction D2 that intersects (orthogonal as an example) the first direction D1. Two of the plurality of third layers 23 in which no optical waveguides 24 are formed are separated from each other in the second direction D2. The two third layers 23 and the two second layers 22 arranged along the second direction D2 have second bonding surfaces 26 that face the first bonding surface 18 of the first semiconductor element 10 described later. The optical waveguides 24 extend in a direction parallel to the second bonding surface 26. The two third layers 23 and the two second layers 22 arranged along the second direction D2 are, for example, support portions 27 that support the first semiconductor element 10. The second semiconductor element 20 has the second bonding surface 26 of the third layer 23 of the support portion 27. The second layer 22 is also referred to as a BOX layer. As an example, the length of the support portion 27 in the second direction D2 is 0.03 mm or more. As an example, the length of the support portion 27 in the third direction D3 is set to a length such that the mesa of the first semiconductor element 10 described later enters between the two third layers 23 arranged along the second direction D2.

[0065] In a top view (viewed from above), the third layer 23 is rectangular. The third layer 23, the second layer 22, and the first layer 21 are arranged in sequence in a third direction D3 that intersects both the first direction D1 and the second direction D2. The third direction D3 is, for example, vertically downward. The second layer 22 is rectangular. In a top view, the third layer 23 may also have the same shape as the second layer 22. Thus, in a top view, the third layer 23 may overlap the second layer 22. For example, the second layer 22 is a SiO2 layer and the third layer 23 is a Si layer. The optical waveguides 24 extend in the first direction D1 in the third layer 23.

[0066] The second semiconductor element 20 has, for example, an electrode 25 provided on the first layer 21. The electrode 25 is made of, for example, gold (Au). The electrode 25 is, for example, on the side opposite to the optical waveguides 24 when viewed from the first semiconductor element 10. The first semiconductor element 10 has an electrode 12 that forms the upper-facing surface of the first semiconductor element 10. The electrode 25 is electrically connected to the electrode 12 via a bonding wire W1, for example. For example, the bonding wire W1 is made of Au.

[0067] The second semiconductor element 20 has a dug-in portion 29 that is recessed in the third direction D3 between the two second layers 22 arranged along the second direction D2. The length (depth) of the dug-in portion 29 in the third direction D3 is, for example, 20 μm or less. It should be noted that the second semiconductor element 20 may not have the dug-in portion 29. The second semiconductor element 20 has, for example, an electrode 28 provided on the dug-in portion 29 between the two second layers 22 arranged along the second direction D2 (see Figure 9 )). The electrode 28 is made of, for example, Au.

[0068] The first semiconductor element 10 is mounted on the third layers 23 among the plurality of third layers 23 where the optical waveguides 24 are not formed. Figure 3 It schematically shows Figure 2 a cross-sectional view of the first semiconductor element 10. As Figure 2 well as Figure 3 shown, the first semiconductor element 10 has the above-described electrodes 12, n-InP layer 13, p-InP layer 14, MQW layer 15, p-InP layer 16, and electrode 17.

[0069] The electrodes 12 and the electrode 17 are made of, for example, gold (Au). The MQW layer 15 contains, for example, a quantum well structure MQW (Multi-Quantum Well). The MQW layer 15 emits an optical signal. The MQW layer 15 is made of, for example, gallium indium arsenide (GaInAs) or gallium indium arsenide phosphide (GaInAsP). The MQW layer 15 may also contain other compound semiconductor materials. The MQW layer 15 and the p-InP layer 16 are formed in a mesa shape on the n-InP layer 13. The p-InP layer 14 has a convex portion 14b that protrudes in the third direction D3. The surface 14d of the convex portion 14b on the side opposite to the mesa is located at a position closer to the center in the second direction D2 of the first semiconductor element 10 than the surface 13d of the n-InP layer 13 on the side opposite to the mesa.

[0070] For example, the length of the first semiconductor element 10 in the first direction D1 is 0.5 mm or more, the length of the first semiconductor element 10 in the second direction D2 is 0.2 mm or more and 0.5 mm or less, and the length of the first semiconductor element 10 in the third direction D3 is 0.05 mm or more. The first semiconductor element 10 has a first bonding surface 18 that is a surface for bonding to the second semiconductor element 20. The aforementioned end surface 11 intersects the first bonding surface 18. The first bonding surface 18 extends along both the first direction D1 and the second direction D2 at the end in the third direction D3 of the convex portion 14b. The first bonding surface 18 is formed, for example, on the p-InP layer 14. The first semiconductor element 10 has a plurality of (two as an example) first bonding surfaces 18 (convex portions 14b). The plurality of first bonding surfaces 18 are arranged along the second direction D2.

[0071] The first bonding surface 18 is formed, for example, by dry etching, wet etching, or epitaxial growth of the p-InP layer 14. Therefore, the flatness of the first bonding surface 18 is good, and it is formed as a surface suitable for hydrophilic bonding described later. As an example, the surface roughness of the first bonding surface 18 is several tens of nm or less. The surface roughness (Ra) of the first bonding surface 18 may be 5 nm or less. The position of the first bonding surface 18 in the third direction D3 is adjusted so that the optical axis position of the optical signal emitted from the MQW layer 15 is aligned with the position of the optical waveguide 24.

[0072] For example, the length of the first joint surface 18 in the second direction D2 is smaller than the length of the support portion 27 in the second direction D2. However, the length of the first joint surface 18 in the second direction D2 may also be larger than the length of the support portion 27 in the second direction D2. When the first semiconductor element 10 is aligned, it moves in the second direction D2. For example, when the length of the support portion 27 in the second direction D2 is larger than the amount of the alignment margin, the joint area of the first joint surface 18 with respect to the second joint surface 26 can be ensured more reliably.

[0073] The first semiconductor element 10 has been described above. However, the configuration of the first semiconductor element is not limited to the aforementioned first semiconductor element 10. Figure 4 FIG. schematically shows a cross section of the first semiconductor element 10A according to the first modification. Hereinafter, a modification of the optical semiconductor device will be described. A part of the configuration of the optical semiconductor device according to the modification is the same as a part of the configuration of the aforementioned optical semiconductor device 1. Therefore, hereinafter, for the descriptions that are repeated with the description of the aforementioned optical semiconductor device 1, the same reference numerals will be given and appropriate omissions will be made.

[0074] The first semiconductor element 10A has an electrode 12, an n-InP layer 13, an MQW layer 15, a p-InP layer 16, and an electrode 17, and does not have a p-InP layer 14. The n-InP layer 13 has a convex portion 13b protruding in the third direction D3. The first semiconductor element 10A has a first joint surface 18A, and the first joint surface 18A is formed on the n-InP layer 13. The first joint surface 18A extends along both the first direction D1 and the second direction D2 at the end of the convex portion 13b in the third direction D3. For example, the shape, size, and number of the first joint surface 18A are the same as those of the aforementioned first joint surface 18. The first joint surface 18A is the interface between the n-InP layer 13 and the MQW layer 15. The first joint surface 18A is formed by dry etching, wet etching, or epitaxial growth of the n-InP layer 13. Therefore, the flatness of the first joint surface 18A is good, and it is formed into a surface suitable for hydrophilic bonding.

[0075] Figure 5 FIG. schematically shows a cross section of the first semiconductor element 10A according to the second modification. In the first semiconductor element 10A according to the second modification, the end portion of the first joint surface 18A on the side opposite to the mesa in the second direction D2 and the surface of the n-InP layer 13 on the side opposite to the mesa are arranged along the third direction D3. That is, the end portion of the first joint surface 18A on the side opposite to the mesa reaches the end portion of the n-InP layer 13 on the side opposite to the mesa. Therefore, in the second modification, the area of the first joint surface 18A can be enlarged compared with the first modification.

[0076] Figure 6 This is a diagram schematically showing a cross-section of the first semiconductor element 10B according to the third modification. The first semiconductor element 10B includes a p-InP layer 14B having a shape different from that of the p-InP layer 14. The p-InP layer 14B has a convex portion 14c with a size different from that of the aforementioned convex portion 14b. The surface 14f of the convex portion 14c facing the opposite side of the mesa is arranged along the third direction D3 with the surface 13d of the n-InP layer 13 facing the opposite side of the mesa. The first semiconductor element 10B has a first bonding surface 18B. The end portion of the first bonding surface 18B in the second direction D2 on the side opposite to the mesa is arranged along the third direction D3 with the end portion of the n-InP layer 13 in the second direction D2. The first bonding surface 18B can be formed in the same manner as the first bonding surface 18, and thus has good flatness and is formed into a surface suitable for hydrophilic bonding.

[0077] Figure 7 This is a diagram schematically showing a cross-section of the first semiconductor element 10C according to the fourth modification. The first semiconductor element 10C includes an electrode 12, an n-InP layer 13, an MQW layer 15, a p-InP layer 16, and an electrode 17. The n-InP layer 13 has a convex portion 13b, and the MQW layer 15 is formed at the end portion of the convex portion 13b in the third direction D3. The first semiconductor element 10C has a first bonding surface 18C, and the first bonding surface 18C is formed on the MQW layer 15. The first bonding surface 18C extends along both the first direction D1 and the second direction D2 at the end portion of the MQW layer 15 formed on the convex portion 13b in the third direction D3. The first bonding surface 18C can be formed in the same manner as the first bonding surface 18, and thus has good flatness and is formed into a surface suitable for hydrophilic bonding.

[0078] Figure 8 This is a diagram schematically showing a cross-section of the first semiconductor element 10C according to the fifth modification. In the first semiconductor element 10C according to the fifth modification, the end portion of the first bonding surface 18C on the side opposite to the mesa is arranged along the third direction D3 with the surface of the n-InP layer 13 facing the opposite side of the mesa. That is, the end portion of the first bonding surface 18C on the side opposite to the mesa reaches the end portion of the n-InP layer 13 on the side opposite to the mesa. Therefore, in the fifth modification, compared with the fourth modification, the area of the first bonding surface 18C can be enlarged.

[0079] Next, a method for manufacturing the optoelectronic device according to the present embodiment will be described. Hereinafter, a method for manufacturing the aforementioned optoelectronic device 1 will be described. First, as Figure 9As shown, a first semiconductor element 10 and a second semiconductor element 20 are prepared. The first semiconductor element 10 has a flat first bonding surface 18. The second semiconductor element 20 has a flat second bonding surface 26. As an example, the surface roughness of the first bonding surface 18 is several tens of nm or less. The surface roughness (Ra) of the first bonding surface 18 can be 5 nm or less. As an example, the surface roughness of the second bonding surface 26 is several tens of nm or less. The surface roughness (Ra) of the second bonding surface 26 can be 5 nm or less. Then, the first bonding surface 18 of the first semiconductor element 10 is hydrophilized (a process of hydrophilizing the first bonding surface). At this time, the second bonding surface 26 of the second semiconductor element 20 can also be hydrophilized. "Hydrophilizing the bonding surface" means, for example, forming hydroxyl groups or silanol groups on the bonding surface. Hereinafter, hydrophilization will sometimes be referred to as the hydrophilization treatment.

[0080] For example, as Figure 10 shown, by hydrophilizing the first bonding surface 18 and the second bonding surface 26, OH groups are formed on the first bonding surface 18 and the second bonding surface 26, respectively. In the hydrophilization treatment, for example, the first bonding surface 18 is hydrophilized by irradiating ultraviolet rays on the first bonding surface 18 in an air-exposed environment. In the hydrophilization treatment, the first bonding surface 18 can also be hydrophilized by exposing the first bonding surface 18 to oxygen plasma in a vacuum environment. In addition, in the hydrophilization treatment, the first bonding surface 18 can also be hydrophilized by exposing the first bonding surface 18 to nitrogen plasma in a vacuum environment. For example, the hydrophilization treatment for the second bonding surface 26 can be performed in the same manner as the hydrophilization treatment for the first bonding surface 18.

[0081] Figure 11 is a diagram schematically showing the steps after the hydrophilization treatment in the manufacturing method of the opto-semiconductor device 1. As Figure 11 shown in (1) of, alignment of the first semiconductor element 10 with respect to the second semiconductor element 20 is performed. At this time, in a state where the first semiconductor element 10 is separated from the second semiconductor element 20, the positions of the first bonding surface 18 in the first direction D1 and the second direction D2 are aligned with the positions of the second bonding surface 26 in the first direction D1 and the second direction D2.

[0082] More specifically, alignment (alignment process) is performed in such a manner that the second bonding surface 26 of the second semiconductor element 20 and the first bonding surface 18 of the first semiconductor element 10 face each other in a separated state, and the end face 11 and the optical waveguide 24 are optically coupled to each other. This alignment is performed in a state where the separation distance between the first bonding surface 18 and the second bonding surface 26 is shortened. For example, the distance between the first bonding surface 18 and the second bonding surface 26 during this alignment is 1 μm or more and less than 100 μm (several μm as an example).

[0083] As Figure 10 and Figure 11 shown, after the above alignment is performed, the first semiconductor element 10 is moved in the third direction D3 to bring the first bonding surface 18 into contact with the second bonding surface 26, and temporary bonding of the first semiconductor element 10 to the second semiconductor element 20 is performed. At this time, the first bonding surface 18 and the second bonding surface 26 are brought into contact at the first temperature, and at least one of the first semiconductor element 10 and the second semiconductor element 20 is pressed to perform temporary bonding to each other (temporary bonding process). For example, the time for temporary bonding is several tens of seconds (10 seconds or less as an example).

[0084] The first temperature is, for example, 20°C or more and 40°C or less. The first temperature can be normal temperature (or room temperature). For example, during temporary bonding, it is not necessary to heat the first bonding surface 18 and the second bonding surface 26. If temporary bonding is performed, the hydrogen atoms of the OH groups on the first bonding surface 18 are bonded to the hydrogen atoms of the OH groups on the second bonding surface 26, and a bonded portion 30 where the first bonding surface 18 and the second bonding surface 26 are bonded to each other is formed. The strength of the bonded portion 30 is, for example, 5 MPa or more.

[0085] After temporary bonding is performed, the first semiconductor element 10 and the second semiconductor element 20 are heated to permanently bond the first bonding surface 18 and the second bonding surface 26 to each other at a second temperature higher than the first temperature (permanent bonding process). The second temperature is, for example, 100°C or more and 300°C or less. As an example, the second temperature can be 150°C. At this time, the bonded portion 30 is heated to cause a hydrolysis reaction in the bonded portion 30. As a result, the OH groups in the bonded portion 30 are decomposed, hydrogen atoms are detached from the bonded portion 30, and the first bonding surface 18 and the second bonding surface 26 are bonded to each other via oxygen atoms. At this time, the first bonding surface 18 and the second bonding surface 26 are hydrophilically bonded, and an oxide film 31 composed of the oxygen atoms is formed in the bonded portion 30. In the hydrophilically bonded state, a bonding bridge of -O- is formed in the bonded portion 30, and thus a state is formed in which the first bonding surface 18 is firmly bonded to the second bonding surface 26. In order to obtain a firm bond based on hydrophilic bonding, it is preferable that the flatness of each of the first bonding surface 18 and the second bonding surface 26 is good.

[0086] After the formal bonding, the first semiconductor element 10 and the second semiconductor element 20 are electrically connected (the process of electrical connection). Specifically, an electrical bonding agent 32 is applied between the electrode 28 of the second semiconductor element 20 and the electrode 17 of the first semiconductor element 10 to electrically connect the electrode 17 and the electrode 28. The electrical bonding agent 32 is, for example, a conductive paste. As an example, the electrical bonding agent 32 is a silver paste. More specifically, after the electrical bonding agent 32 is applied, the first semiconductor element 10 and the second semiconductor element 20 are heated at a third temperature higher than the first temperature to bond the electrode 17 and the electrode 28 to each other. The third temperature is set to a temperature at which the electrical bonding agent 32 is formed into a molten state. The third temperature is, for example, set to be higher than the melting point of a single metal or alloy contained in the electrical bonding agent 32. When the electrical bonding agent 32 is solder, the third temperature is set to be equal to or higher than the melting point of the solder. The third temperature is, for example, 120°C or higher and 350°C or lower. As an example, the third temperature can be 190°C. After the electrode 17 and the electrode 28 are connected to each other, the electrode 12 of the first semiconductor element 10 and the electrode 25 of the second semiconductor element 20 are connected via a bonding wire W1. After the first semiconductor element 10 and the second semiconductor element 20 are electrically connected as described above, a series of processes of the manufacturing method of the optical semiconductor device 1 according to the present embodiment are completed.

[0087] Next, the effects obtained by the optical semiconductor device 1 and the manufacturing method of the optical semiconductor device 1 according to the present embodiment will be described. In the optical semiconductor device 1 and the manufacturing method of the optical semiconductor device 1, the first semiconductor element 10 has a first bonding surface 18 and an end surface 11, and the second semiconductor element 20 has a second bonding surface 26 and an optical waveguide 24. The end surface 11 of the first semiconductor element 10 and the optical waveguide 24 of the second semiconductor element 20 are optically coupled to each other, and the first bonding surface 18 and the second bonding surface 26 are hydrophilically bonded to each other. In the hydrophilic bonding, heating after alignment of the first bonding surface 18 with respect to the second bonding surface 26 is not required. In the present embodiment, at the time of temporary bonding, the first bonding surface 18 of the first semiconductor element 10 is fixed to the second bonding surface 26 of the second semiconductor element 20. Therefore, it is possible to suppress the case where the position of the first semiconductor element 10 with respect to the second semiconductor element 20 is shifted due to heating. Therefore, it is possible to suppress the position shift and mount the first semiconductor element 10 with respect to the second semiconductor element 20 with high precision. As a result, it is possible to suppress the case where the efficiency of optical coupling caused by the alignment of the end surface 11 of the first semiconductor element 10 and the optical waveguide 24 of the second semiconductor element 20 is reduced after bonding.

[0088] As described above, an oxide film 31 may also be formed at a joint portion 30 where the first joint surface 18 and the second joint surface 26 are hydrophilically joined to each other. In this case, the first joint surface 18 and the second joint surface 26 can be firmly joined by oxygen atoms constituting the oxide film 31.

[0089] As described above, in the hydrophilization process, the first joint surface 18 can be hydrophilized by irradiating ultraviolet rays on the first joint surface 18 in an air-exposed environment. In this case, it is possible to suppress damage to the first joint surface 18 when hydrophilizing the first joint surface 18. Thereby, it is possible to reduce the deterioration of the hydrophilically joined state due to the roughening of the first joint surface. For example, the thickness of the oxide film 31 is 10 nm or less.

[0090] As described above, in the hydrophilization process, the first joint surface 18 can also be hydrophilized by exposing the first joint surface 18 to oxygen plasma in a vacuum environment. In this case, when hydrophilizing the first joint surface 18, a thick oxide film 31 can be formed on the first joint surface 18. The joining force can be enhanced by the thick oxide film. For example, the thickness of the oxide film 31 is 10 nm or more.

[0091] As described above, in the hydrophilization process, the first joint surface 18 can also be hydrophilized by exposing the first joint surface 18 to nitrogen plasma in a vacuum environment.

[0092] As described above, the strength of the joint portion 30, which is the portion where the first joint surface 18 and the second joint surface 26 are joined to each other after the temporary joining process, can be 5 MPa or more. In this case, the first semiconductor element 10 can be firmly joined to the second semiconductor element 20 in the temporary joining.

[0093] As described above, the distance between the first joint surface 18 and the second joint surface 26 during the alignment process can be 1 μm or more and less than 100 μm. In this case, the distance between the first joint surface 18 and the second joint surface 26 during alignment can be reduced, so that the positional deviation of the first joint surface 18 with respect to the second joint surface 26 can be more reliably suppressed.

[0094] As described above, the first temperature can be 20°C or more and 40°C or less, and the second temperature can be 100°C or more and 300°C or less. In this case, the first temperature during the temporary joining can be set to room temperature, so that the positional deviation accompanied by heating can be more reliably suppressed.

[0095] Next, refer to Figure 12 、 Figure 13 and Figure 14The opto-semiconductor device 1A according to the second embodiment will be described. The opto-semiconductor device 1A includes a first semiconductor element 40 and a second semiconductor element 50. The first semiconductor element 40 has an end face 11 in the same manner as the aforementioned first semiconductor element 10. The second semiconductor element 50 has a second bonding surface 56 and a support portion 57 whose lengths in the second direction D2 are smaller than those of the second bonding surface 26 and the support portion 27.

[0096] The first semiconductor element 40 includes an electrode 12, an n-InP layer 13, a p-InP layer 44, an MQW layer 15, a p-InP layer 46, an electrode 17, and an n-InP layer 47. The p-InP layer 44 has a recess 44b that is recessed in a direction opposite to the third direction D3. The recess 44b is defined by a first side portion 44c in contact with the MQW layer 15, a bottom portion 44d extending from the first side portion 44c in a direction opposite to the MQW layer 15, and a second side portion 44f protruding in the third direction D3 at an end of the bottom portion 44d on a side opposite to the first side portion 44c. n-InP layers 47 are formed at ends of the first side portion 44c and the second side portion 44f in the third direction D3. A p-InP layer 46 is formed at an end of the n-InP layer 47 in the third direction D3. The n-InP layer 47 is provided to suppress current from flowing from the p-InP layer 46 to the p-InP layer 44.

[0097] As Figure 15 shown in the modification example of, the first semiconductor element 40 may not have Figure 14 the second side portion 44f shown, the n-InP layer 47 formed on the second side portion 44f, and the p-InP layer 46 formed on the n-InP layer 47. In this way, the shape of the first semiconductor element 40 can be appropriately changed. The first semiconductor element 40 has a first bonding surface 48 at an end of the bottom portion 44d in the third direction D3. The first bonding surface 48 is formed by etching. In this case, the position of the first bonding surface 48 in the third direction D3 can be adjusted by the etching amount.

[0098] For example, the position of the first bonding surface 48 in the third direction D3 can also be set to the position of the optical axis of the optical signal emitted from the MQW layer 15 in the third direction D3. The length of the first bonding surface 48 in the second direction D2 is greater than the length of the support portion 57 in the second direction D2. The first semiconductor element 40 moves in the second direction D2 when being aligned. For example, when the length of the first bonding surface 48 in the second direction D2 is greater than the amount of the alignment margin by that amount, the bonding area of the second bonding surface 56 with respect to the first bonding surface 48 can be more reliably ensured.

[0099] Figure 16FIG. is a cross-sectional view schematically showing a first semiconductor element 40A according to another modification. The first semiconductor element 40A is different from the aforementioned first semiconductor element 40 in that a bonding layer 45 is formed at the bottom 44d of the recess 44b. For example, the material of the bonding layer 45 is the same as the material of the MQW layer 15. The first semiconductor element 40A has a first bonding surface 48A, and the first bonding surface 48A is formed on the bonding layer 45. The first bonding surface 48A extends along both the first direction D1 and the second direction D2 at the end in the third direction D3 of the bonding layer 45. The first bonding surface 48A is formed by etching, for example. The first bonding surface 48A may also be formed by dry etching, wet etching, or epitaxial growth together with the MQW layer 15. In this case, the flatness of the first bonding surface 48A can be made good, and the first bonding surface 48A can be formed into a surface suitable for hydrophilic bonding.

[0100] As Figure 17 shown in the modification of, the first semiconductor element 40A may not have Figure 16 the second side portion 44f shown, the n-InP layer 47 formed on the second side portion 44f, and the p-InP layer 46 formed on the n-InP layer 47. In Figure 17 the first semiconductor element 40A shown, the end portion of the first bonding surface 48A on the side opposite to the mesa and the end portion of the p-InP layer 44 on the side opposite to the mesa are arranged along the third direction D3. That is, the end portion of the first bonding surface 48A on the side opposite to the mesa reaches the end portion of the p-InP layer 44 on the side opposite to the mesa. Therefore, in Figure 17 the first semiconductor element 40A according to the modification of, compared with Figure 16 the first semiconductor element 40A of, the area of the first bonding surface 48A can be enlarged.

[0101] Next, a method for manufacturing the optoelectronic device 1A will be described. A part of the process of the method for manufacturing the optoelectronic device 1A is the same as a part of the process of the method for manufacturing the aforementioned optoelectronic device 1, so the repeated description will be appropriately omitted. Hereinafter, a method for manufacturing the optoelectronic device 1A having the first semiconductor element 40 will be described. It should be noted that the method for manufacturing the optoelectronic device 1A having the first semiconductor element 40A is the same as the method for manufacturing the optoelectronic device 1A having the first semiconductor element 40, so the description will be omitted.

[0102] First, prepare a first semiconductor element 40 and a second semiconductor element 50, and perform a hydrophilic treatment on the first bonding surface 48 of the first semiconductor element 40 (a process of hydrophilizing the first bonding surface). The hydrophilic treatment is performed, in the same manner as in the first embodiment, by irradiating ultraviolet rays on the first bonding surface 48 in an air-exposed environment, or by exposing it to oxygen plasma or nitrogen plasma in a vacuum environment. At this time, the second bonding surface 56 of the second semiconductor element 50 may also be hydrophilized. At this time, OH groups are formed on the first bonding surface 48 and the second bonding surface 56, respectively.

[0103] Figure 18 is a diagram schematically showing the steps of a method for manufacturing the opto-semiconductor device 1A. As Figure 18 shown in (1) of, alignment of the first semiconductor element 40 with respect to the second semiconductor element 50 is performed. At this time, in a state where the first semiconductor element 40 is separated from the second semiconductor element 50, the positions of the first direction D1 and the second direction D2 of the concave portion 44b are aligned with the positions of the first direction D1 and the second direction D2 of the second bonding surface 56. Then, alignment is performed in such a manner that the end face 11 and the optical waveguide 24 are optically coupled to each other (a process of performing alignment).

[0104] After alignment is performed, temporary bonding and permanent bonding are performed in the same manner as in the first embodiment. Specifically, the first semiconductor element 40 is moved in the third direction D3 so that the first bonding surface 48 comes into contact with the second bonding surface 56, and at least one of the first semiconductor element 40 and the second semiconductor element 50 is pressed in an environment at the first temperature to perform temporary bonding with each other (a process of performing temporary bonding). At this time, the hydrogen atoms of the OH groups on the first bonding surface 48 are bonded to the hydrogen atoms of the OH groups on the second bonding surface 56, and a bonding portion 30A, which is a portion where the first bonding surface 48 and the second bonding surface 56 are bonded to each other, is formed.

[0105] After temporary bonding is performed, the first semiconductor element 40 and the second semiconductor element 50 are heated in an environment at the second temperature to perform permanent bonding (a process of performing permanent bonding). At this time, the OH groups in the bonding portion 30A are decomposed, the hydrogen atoms are detached from the bonding portion 30A, and the first bonding surface 48 and the second bonding surface 56 are bonded to each other via oxygen atoms. The first bonding surface 48 and the second bonding surface 56 are hydrophilically bonded to each other, and an oxide film 31A composed of oxygen atoms is formed in the bonding portion 30A.

[0106] After the formal bonding, the first semiconductor element 40 and the second semiconductor element 50 are electrically connected (the process of electrical connection). Specifically, an electrical bonding agent 32 is applied between the electrode 28 of the second semiconductor element 50 and the electrode 17 of the first semiconductor element 40 to electrically connect the electrode 17 and the electrode 28. The electrical bonding agent 32 is, for example, a conductive paste. As an example, the electrical bonding agent 32 is a silver paste. More specifically, after the electrical bonding agent 32 is applied, the first semiconductor element 40 and the second semiconductor element 50 are heated at a third temperature higher than the first temperature to bond the electrode 17 and the electrode 28 to each other. The third temperature is, for example, 120 °C or higher and 350 °C or lower. As an example, the third temperature can be 190 °C. After the electrode 17 and the electrode 28 are connected to each other, the electrode 12 of the first semiconductor element 40 and the electrode 25 of the second semiconductor element 50 are connected via a bonding wire W1. After the first semiconductor element 40 and the second semiconductor element 50 are electrically connected as described above, a series of processes of the manufacturing method of the optical semiconductor device 1A are completed.

[0107] As described above, in the optical semiconductor device 1A and the manufacturing method of the optical semiconductor device 1A according to the second embodiment, the first bonding surface 48 is hydrophilized, and the first semiconductor element 40 is temporarily bonded to the second semiconductor element 50 in an environment of the first temperature. In the temporary bonding, the first bonding surface 48 is fixed to the second bonding surface 56. Therefore, it is possible to suppress the displacement of the position of the first semiconductor element 40 relative to the second semiconductor element 50 due to heating, and thus the same effects as those of the aforementioned optical semiconductor device 1 can be obtained. In addition, in the second embodiment, the first semiconductor element 40 has a concave portion 44b, and the first bonding surface 48 is formed at the bottom 44d of the concave portion 44b, whereby the alignment of the first bonding surface 48 with respect to the second bonding surface 56 can be more easily performed.

[0108] Next, with reference to Figure 19 and Figure 20 the optical semiconductor device 1B and the manufacturing method of the optical semiconductor device 1B according to the third embodiment will be described. Figure 20 The figure of the back surface of is Figure 20An enlarged view of a portion between a pair of support portions 27 on the upper surface. The opto-semiconductor device 1B includes a first semiconductor element 60 and a second semiconductor element 70. The first semiconductor element 60 is different from the first semiconductor element 10 in that it does not have the aforementioned electrode 12 and has two electrodes 17. One of the two electrodes 17 has an extension portion 17b that penetrates upward through the p-InP layer 16 and the MQW layer 15 and reaches the n-InP layer 13. The second semiconductor element 70 is different from the second semiconductor element 20 in that it does not have the aforementioned electrode 25 and has two electrodes 28. In the opto-semiconductor device 1B, the bonding wire W1 is not required. For example, the distance L from the end face 11 to the optical waveguide 24 is 50 μm or less.

[0109] A method for manufacturing the opto-semiconductor device 1B will be described. The steps of hydrophilizing the first bonding surface 18, aligning, temporarily bonding, and permanently bonding are the same as those in the first embodiment. In the third embodiment, an electrical bonding agent 32 is applied between each of the two electrodes 28 of the second semiconductor element 70 and each of the two electrodes 17 of the first semiconductor element 60 for electrical connection. After electrically connecting the first semiconductor element 60 and the second semiconductor element 70 in this way, a series of steps of the manufacturing method of the opto-semiconductor device 1B is completed. According to the opto-semiconductor device 1B and the manufacturing method of the opto-semiconductor device 1B according to the third embodiment, the same effects as those of the first embodiment can be obtained, and the effect of not requiring wire bonding can be achieved.

[0110] Refer to Figure 21 An opto-semiconductor device 1C and a manufacturing method of the opto-semiconductor device 1C according to the fourth embodiment will be described. The opto-semiconductor device 1C includes a first semiconductor element 80 and a second semiconductor element 90. As Figure 21 shown in (4) of, the first semiconductor element 80 is different from the first semiconductor element 10 in that it has two electrodes 12, and the second semiconductor element 90 is different from the second semiconductor element 20 in that it has two electrodes 25. The two electrodes 12 are separated from each other. The two electrodes 12 are, for example, an n electrode and a p electrode. For example, the two electrodes 12 are arranged along the second direction D2, and the two electrodes 25 are arranged along the second direction D2.

[0111] In the manufacturing method of the opto-semiconductor device 1C, the steps of hydrophilizing the first bonding surface 18, aligning, temporarily bonding, and permanently bonding are the same as those in the first embodiment. In the fourth embodiment, the two electrodes 25 of the second semiconductor element 90 are respectively connected to the two electrodes 12 of the first semiconductor element 80 via the bonding wire W1. After electrically connecting the first semiconductor element 80 and the second semiconductor element 90 in this way, a series of steps of the manufacturing method of the opto-semiconductor device 1C is completed.

[0112] Next, with reference to Figure 22 the manufacturing method of the opto-semiconductor device 1 according to the fifth embodiment will be described. As Figure 22 shown, in the manufacturing method of the opto-semiconductor device 1 according to the fifth embodiment, the step of hydrophilizing the first bonding surface 18, the step of alignment, and the step of temporary bonding are the same as those in the first embodiment. In the fifth embodiment, after temporary fixing and before permanent fixing, an electrical bonding agent 32 is applied (the step of applying the electrical bonding agent). At this time, the electrical bonding agent 32 is applied between the electrode 28 and the electrode 17.

[0113] After the electrical bonding agent 32 is applied, permanent bonding is performed. The method of permanent bonding is the same as that in the first embodiment. At this time, the electrical bonding agent 32 is baked while baking for permanent bonding. Then, the electrode 25 of the second semiconductor element 20 is connected to the electrode 12 of the first semiconductor element 10 via the bonding wire W1. After electrically connecting the first semiconductor element 10 and the second semiconductor element 20 in this way, a series of steps of the manufacturing method of the opto-semiconductor device 1 are completed. According to the manufacturing method of the opto-semiconductor device 1 according to the fifth embodiment, the same effects as those in the first embodiment can be obtained, and the effect of simplifying the process can be obtained because the electrical bonding agent 32 can be baked while baking for permanent bonding.

[0114] With reference to Figure 23 the opto-semiconductor device 1D according to the sixth embodiment will be described. The opto-semiconductor device 1D includes a first semiconductor element 110 and a second semiconductor element 120. The second semiconductor element 120 has the same first layer 21, second layer 22 located above the first layer 21, and third layer 23 located above the second layer 22 as the aforementioned second semiconductor element 20. The second layer 22 extends along a first direction D1 which is the direction in which the optical waveguide 24 extends and a second direction D2 which intersects the first direction D1. Two third layers 23 and two second layers 22 arranged along the second direction D2 are support portions 27 for supporting the first semiconductor element 110. The second semiconductor element 120 has a second bonding surface 26 which is the upper surface of the support portion 27. The second semiconductor element 120 has two second bonding surfaces 26, and the two second bonding surfaces 26 are arranged along the second direction D2.

[0115] The second semiconductor element 120 has an electrode 125. The electrode 125 is provided, for example, on the first layer 21. The electrode 125 is made of, for example, any one of gold (Au), titanium (Ti), and platinum (Pt). The electrode 125 is located, for example, at an end portion on the opposite side of the first direction D1 of the first layer 21 and on the opposite side of the second direction D2 of the first layer 21 when the second semiconductor element 120 is viewed from above (when the second semiconductor element 120 is viewed from above). The second semiconductor element 120 may also have a plurality of electrodes 125 (see Figure 25 ). In this case, when viewed from above the second semiconductor element 120, on the side of the first layer 21 opposite to the first direction D1, the two electrodes 125 may be arranged along the second direction D2.

[0116] Before the first semiconductor element 110 and the second semiconductor element 120 are joined, the second semiconductor element 120 has a first bonding layer 126 made of an electrical bonding agent and a second bonding layer 127 formed on the first bonding layer 126 and made of an electrical bonding agent. For example, the first bonding layer 126 and the second bonding layer 127 are each solder. As an example, the material of the first bonding layer 126 and the material of the second bonding layer 127 are gold-tin (AuSn). However, the material of the first bonding layer 126 and the material of the second bonding layer 127 may also be tin-silver-copper (SnAgCu), and there is no particular limitation. The first bonding layer 126 and the second bonding layer 127 melt when heated. After melting, the second bonding layer 127 on the first bonding layer 126 flows into a portion of the first bonding layer 126 where the second bonding layer 127 is not formed (above the portion of the first bonding layer 126 exposed before heating, above the first bonding layer region 126A described later).

[0117] Figure 24 is a perspective view showing an enlarged view of a part of the first bonding layer 126 and a part of the second bonding layer 127. As shown in Figure 23 and Figure 24 , the second semiconductor element 120 has a flow stopper 128 located between the first bonding layer 126 and the second bonding layer 127 and the electrode 125. The flow stopper 128 prevents the melted solder (the first bonding layer 126 and the second bonding layer 127) from flowing into the electrode 125. The flow stopper 128 is made of, for example, Pt. For example, at a position facing the electrode 125, the first bonding layer 126 and the second bonding layer 127 are arranged along the first direction D1. The flow stopper 128 extends along the first direction D1 between the portion where the first bonding layer 126 and the second bonding layer 127 are arranged and the electrode 125.

[0118] Figure 25is a perspective view showing the second semiconductor element 120. As Figure 25 shown, before the first semiconductor element 110 and the second semiconductor element 120 are joined, the second semiconductor element 120 has a first bonding agent layer region 126A in which only the first bonding agent layer 126 is formed, and a region in which the second bonding agent layer 127 is formed on the first bonding agent layer 126, that is, a second bonding agent layer region 127A. For example, the second semiconductor element 120 has one first bonding agent layer region 126A and a plurality of second bonding agent layer regions 127A. The first bonding agent layer region 126A includes, for example, a first portion 126b located between two support portions 27 arranged along the second direction D2, a second portion 126c located at an end of the first portion 126b in the first direction D1, and a third portion 126d located at an end of the first portion 126b in a direction opposite to the first direction D1.

[0119] The first portion 126b is formed in a rectangular shape extending along the first direction D1. The length of the first portion 126b in the first direction D1 is greater than the length of the support portion 27 in the first direction D1. The end of the first portion 126b in the first direction D1 protrudes in the first direction D1 from the end of the support portion 27 in the first direction D1. The end of the first portion 126b in a direction opposite to the first direction D1 protrudes in the opposite direction from the end of the support portion 27 in the opposite direction. The second portion 126c is formed in a rectangular shape extending along the second direction D2. The end of the second portion 126c in the second direction D2 protrudes in the second direction D2 from the end of the first portion 126b in the second direction D2. The end of the second portion 126c in a direction opposite to the second direction D2 protrudes in the opposite direction from the end of the first portion 126b in the opposite direction.

[0120] When the second semiconductor element 120 is viewed from above, the first portion 126b and the second portion 126c are formed in a T shape. The third portion 126d is formed in a rectangular shape extending along the second direction D2. The third portion 126d protrudes from the first portion 126b in a direction opposite to the second direction D2. When the second semiconductor element 120 is viewed from above, for example, the first portion 126b and the third portion 126d are formed in an L shape. A flow stop member 128 is formed in a direction opposite to the second direction D2 of the third portion 126d.

[0121] The second adhesive layer region 127A includes, for example, a first portion 127b and a second portion 127c that are located in the first direction D1 with respect to the support portion 27, and a third portion 127d that is located in the direction opposite to the first direction D1 with respect to the first adhesive layer region 126A. The first portion 127b is located in the first direction D1 with respect to the support portion 27 among the two support portions 27 that is located in the second direction D2. The first portion 127b is located in the second direction D2 with respect to the second portion 126c. The first portion 127b is formed in a rectangular shape, for example.

[0122] The second portion 127c is located in the first direction D1 with respect to the support portion 27 among the two support portions 27 that is located in the direction opposite to the second direction D2. The second portion 127c and the first portion 127b are arranged along the second direction D2. The second portion 127c and the first portion 127b sandwich the second portion 126c along the second direction D2. The second portion 127c is formed in a rectangular shape. The third portion 127d is formed in a rectangular shape that extends along the second direction D2. The third portion 127d is located in the direction opposite to the first direction D1 with respect to the third portion 126d of the first adhesive layer region 126A. A flow stopper 128 is formed in the direction opposite to the second direction D2 of the third portion 127d.

[0123] Figure 26 FIG. is a cross-sectional view and a partial enlarged view when the first adhesive layer 126 (first portion 126b) and the second adhesive layer 127 (third portion 127d) are cut along a plane extending in the first direction D1 and the third direction D3. Figure 27 FIG. is a cross-sectional view and a partial enlarged view when the first adhesive layer 126 (second portion 126c) and the second adhesive layer 127 (first portion 127b and second portion 127c) are cut along a plane extending in the second direction D2 and the third direction D3. As Figure 26 and Figure 27 shown, for example, the thickness T2 of the second adhesive layer 127 is larger than the thickness T1 of the first adhesive layer 126.

[0124] Figure 28 FIG. represents a cross-section when the first semiconductor element 110 and the second semiconductor element 120 are cut along a plane extending in the second direction D2 and the third direction D3, showing the states before and after the first adhesive layer 126 and the second adhesive layer 127 are melted. As Figure 26 , Figure 27 and Figure 28 shown, after melting, the first portion 127b and the second portion 127c flow into the second portion 126c, and the melted third portion 127d flows into the third portion 126d and the first portion 126b.

[0125] Figure 29 This shows a cross-section when the first semiconductor element 110 and the second semiconductor element 120 are cut along a plane extending in the first direction D1 and the third direction D3, showing the situation before and after the first bonding agent layer 126 and the second bonding agent layer 127 are melted. As Figure 27 、 Figure 28 and Figure 29 shown, the thickness T1 of the first bonding agent layer 126 is set to a thickness that does not contact the electrode 17 of the first semiconductor element 110 when the second bonding surface 26 of the second semiconductor element 120 is bonded to the first bonding surface 18 of the first semiconductor element 110.

[0126] The thickness T2 of the second bonding agent layer 127 is set to the height at which the bonding agent layer 130 contacts the third bonding surface 131 when the first bonding agent layer 126 and the second bonding agent layer 127 are melted. The third bonding surface 131 is parallel to the first bonding surface 18. The third bonding surface 131 is, for example, the aforementioned electrode 17. The second semiconductor element 120 has a fourth bonding surface 132 facing the third bonding surface 131. The fourth bonding surface 132 is, for example, the electrode 28 of the second semiconductor element 120 (refer to Figure 30 ).

[0127] For example, when the height of the gap S between the third bonding surface 131 and the fourth bonding surface 132 before the first bonding agent layer 126 and the second bonding agent layer 127 are melted is set to D, the sum of the thickness T1 and the thickness T2 is greater than D. As an example, D is 3 μm or more and 4 μm or less. For example, the thickness T1 is less than the thickness T2. For example, the thickness T1 is 0.1 μm or more and 20 μm or less, and the thickness T2 is 0.1 μm or more and 50 μm or less. As an example, the thickness T1 is 2 μm and the thickness T2 is 5 μm.

[0128] In a state where the first bonding agent layer 126 and the second bonding agent layer 127 are melted, the opto-semiconductor device 1D includes a bonding agent layer 130 that bonds the third bonding surface 131 and the fourth bonding surface 132 to each other at the bonding portion. The area of the bonding agent layer 130 when observed along the direction (crossing direction) crossing the third bonding surface 131, that is, the third direction D3, is larger than the area of the bonding portion when observed along the third direction D3. For example, the area of the bonding agent layer 130 when observed along the third direction D3 is equal to or less than the area of the third bonding surface 131 or the fourth bonding surface 132 when observed along the third direction D3.

[0129] For example, the area of the third joint surface 131 when observed along the third direction D3 is equal to the area of the fourth joint surface 132 when observed along the third direction D3. The area ratio of the adhesive layer 130 when observed along the third direction D3 to the area of the third joint surface 131 when observed along the third direction D3 is, for example, 1.1 or more and 5.0 or less.

[0130] Next, with reference to Figure 29 and Figure 30 an example of the manufacturing method of the opto-semiconductor device 1D will be described. Figure 30 is a diagram schematically showing the processes after the hydrophilic treatment in the manufacturing method of the opto-semiconductor device 1D. Hereinafter, descriptions overlapping with those of the manufacturing method of the aforementioned opto-semiconductor device 1 will be appropriately omitted. As Figure 29 and Figure 30 shown in (1) of, a first adhesive layer 126 is formed on the fourth joint surface 132 of the second semiconductor element 120, and a second adhesive layer 127 is formed on the first adhesive layer 126 (the process of forming the second adhesive layer on the first adhesive layer). The formation of the first adhesive layer 126 and the formation of the second adhesive layer 127 are performed, for example, by vapor deposition.

[0131] Then, alignment of the first semiconductor element 110 with respect to the second semiconductor element 120 is performed. At this time, the second joint surface 26 of the second semiconductor element 120 is opposed to the first joint surface 18 of the first semiconductor element 110 in a separated state for alignment. As Figure 30 shown in (2) of, the first semiconductor element 110 is moved in the third direction D3 to bring the first joint surface 18 into contact with the second joint surface 26, and temporary bonding of the first semiconductor element 110 with respect to the second semiconductor element 120 is performed. After the temporary bonding is performed, the first semiconductor element 110 and the second semiconductor element 120 are heated to formally bond the first joint surface 18 and the second joint surface 26 to each other. The methods of temporary bonding and formal bonding in the manufacturing method of the opto-semiconductor device 1D are the same as the methods of temporary bonding and formal bonding in the manufacturing method of the aforementioned opto-semiconductor device 1.

[0132] As Figure 29 and Figure 30As shown in (3) thereof, by heating the first bonding agent layer 126 and the second bonding agent layer 127, the first bonding agent layer 126 and the second bonding agent layer 127 are melted. Then, the third bonding surface 131 and the fourth bonding surface 132 are bonded to each other by the melted first bonding agent layer 126 and second bonding agent layer 127, i.e., the bonding agent layer 130 (the process of bonding to each other). The heating of the first bonding agent layer 126 and the second bonding agent layer 127 is performed, for example, by reflow soldering. However, the heating of the first bonding agent layer 126 and the second bonding agent layer 127 may also be performed by a heater, and the method of heating the first bonding agent layer 126 and the second bonding agent layer 127 is not particularly limited.

[0133] The heating of the first bonding agent layer 126 and the second bonding agent layer 127 is performed, for example, after the aforementioned formal bonding. The heating of the first bonding agent layer 126 and the second bonding agent layer 127 may be performed after setting to room temperature after the formal bonding, or the heating of the first bonding agent layer 126 and the second bonding agent layer 127 may be performed without setting to room temperature after the formal bonding. The heating temperature of the first bonding agent layer 126 and the second bonding agent layer 127 is higher than the aforementioned second temperature during the formal bonding. The heating temperature of the first bonding agent layer 126 and the second bonding agent layer 127 is set to the temperature at which the first bonding agent layer 126 and the second bonding agent layer 127 become a molten state. The heating temperature is set, for example, higher than the melting point of a single metal or alloy contained in the first bonding agent layer 126 and the second bonding agent layer 127. When the electrical bonding agent 32 is solder, the heating temperature is set to be above the melting point of the solder. The third temperature is, for example, 120 °C or higher and 350 °C or lower (280 °C as an example). The heating time of the first bonding agent layer 126 and the second bonding agent layer 127 is, for example, 1 second or longer and 300 seconds or shorter. As described above, when the first bonding agent layer 126 and the second bonding agent layer 127 are heated, the first bonding agent layer 126 and the second bonding agent layer 127 are melted, and a state is formed in which the bonding agent layer 130 enters the gap S.

[0134] Thereby, a state is formed in which the third bonding surface 131 and the fourth bonding surface 132 are bonded to each other via the bonding agent layer 130. After bonding the third bonding surface 131 and the fourth bonding surface 132 to each other via the bonding agent layer 130, as Figure 30 shown in (4) thereof, the electrode 12 of the first semiconductor element 110 is connected to the electrode 125 of the second semiconductor element 120 via the bonding wire W1. After electrically connecting the first semiconductor element 110 and the second semiconductor element 120 as described above, a series of processes of the manufacturing method of the optical semiconductor device 1D are completed.

[0135] As described above, in the optical semiconductor device 1D according to the sixth embodiment, the first semiconductor element 110 has a third bonding surface 131 parallel to the first bonding surface 18, and the second semiconductor element 120 has a fourth bonding surface 132 facing the third bonding surface 131. The optical semiconductor device 1D further includes an adhesive layer 130 that bonds the third bonding surface 131 and the fourth bonding surface 132 to each other at the bonding portion. The area of the adhesive layer 130 as viewed in the direction intersecting the third bonding surface 131, i.e., the intersecting direction, is larger than the area of the bonding portion as viewed in the intersecting direction. In this case, on the basis of the state in which the adhesive layer 130 is filled between the third bonding surface 131 and the fourth bonding surface 132 at the bonding portion, by making the area of the adhesive layer 130 larger than the area of the bonding portion, the third bonding surface 131 and the fourth bonding surface 132 can be reliably bonded. For example, at the bonding portion, the third bonding surface 131 and the fourth bonding surface 132 can be evenly bonded. As a result, the stress generated in the direction within the plane parallel to the first direction D1 and the second direction D2 due to uneven bonding is reduced, and the positional deviation when bonding the electrode 17 of the first semiconductor element 110 and the electrode 28 of the second semiconductor element 120 to each other can be reduced.

[0136] The manufacturing method of the optical semiconductor device 1D includes the steps of forming a first adhesive layer 126 made of an electrical adhesive on the fourth bonding surface 132 and forming a second adhesive layer 127 made of an electrical adhesive on the first adhesive layer 126 before the alignment process. The manufacturing method includes the steps of melting the first adhesive layer 126 and the second adhesive layer 127 by heating, and bonding the third bonding surface 131 and the fourth bonding surface 132 to each other through the melted first adhesive layer 126 and the second adhesive layer 127, i.e., the adhesive layer 130.

[0137] Since the melted adhesive layer 130 is filled between the third bonding surface 131 and the fourth bonding surface 132, the first semiconductor element 110 and the second semiconductor element 120 can be connected with a small positional deviation. In addition, the heat generated in the first semiconductor element can be efficiently dissipated to the second semiconductor element through the adhesive layer 130, and the operation of the first semiconductor element can be stabilized.

[0138] Next, the optical semiconductor device 1E according to the seventh embodiment will be described. Figure 31 It is a perspective view showing the optical semiconductor device 1E. Figure 32 It is a perspective view showing the second semiconductor element 140 of the optical semiconductor device 1E. As Figure 31 and Figure 32As shown, the opto-semiconductor device 1E includes a first semiconductor element 110 and a second semiconductor element 140. The second semiconductor element 140 is different from the aforementioned second semiconductor element 120 in that it has a second bonding surface 146 different from the second bonding surface 26 and a support portion 147 different from the support portion 27. For example, the four support portions 147 are arranged in a quadrilateral shape.

[0139] The second bonding surface 146 has a first small bonding surface 146b and a second small bonding surface 146c that are separated from each other along a direction parallel to the second bonding surface 146, i.e., the first direction D1. The second bonding surface 146 has two first small bonding surfaces 146b, and the two first small bonding surfaces 146b are arranged along the second direction D2. The second bonding surface 146 has two second small bonding surfaces 146c, and the two second small bonding surfaces 146c are arranged along the second direction D2. When the second semiconductor element 140 is viewed from above, the two first small bonding surfaces 146b and the two second small bonding surfaces 146c are arranged in a rectangular shape.

[0140] The second semiconductor element 140 has an electrode 145. The electrode 145 includes: a first electrode 145b that extends from between the two first small bonding surfaces 146b arranged along the second direction D2 in a direction opposite to the first direction D1; and a second electrode 145c that extends from between the two second small bonding surfaces 146c arranged along the second direction D2 in the first direction D1. The first electrode 145b includes a first portion 145d that extends along the first direction D1 between the two first small bonding surfaces 146b, and a second portion 145f that extends from an end of the first portion 145d in a direction opposite to the first direction D1 in a direction opposite to the second direction D2. The second electrode 145c includes a first portion 145h that extends along the first direction D1 between the two second small bonding surfaces 146c and a second portion 145j that extends along the second direction D2 at an end of the first portion 145h in the first direction D1. The second portion 145j protrudes in two directions, i.e., the second direction D2 and a direction opposite to the second direction D2, from an end of the second portion 145j in the first direction D1.

[0141] The second semiconductor element 140 has a first bonding agent layer 126 made of an electrically conductive bonding agent before the first semiconductor element 110 and the second semiconductor element 140 are bonded, and a second bonding agent layer 127 formed on the first bonding agent layer 126 and made of an electrically conductive bonding agent. Figure 33 It is a cross-sectional view of the second semiconductor element 140 when cut along a plane extending in the first direction D1 and the third direction D3. Figure 34 It is Figure 33 a view obtained by magnifying a portion between the first small bonding surface 146b and the second small bonding surface 146c in the cross-sectional view. As Figure 32, Figure 33 and Figure 34 As shown, the second semiconductor element 140 has a first bonding agent layer 126 located in the first direction D1 and a flow stopper 148 between the second bonding agent layer 127 and the electrode 145.

[0142] The flow stopper 148 is formed, for example, by etching. Through the flow stopper 148, the molten solder (the first bonding agent layer 126 and the second bonding agent layer 127) is prevented from flowing into the electrode 145. The material of the flow stopper 148 is, for example, the same as the material of the aforementioned flow stopper 128. The flow stoppers 148 are respectively formed at the boundary portions between the first bonding agent layer 126 and the first electrode 145b, and between the first bonding agent layer 126 and the second electrode 145c. The flow stopper 148 extends along the second direction D2. The second semiconductor element 140 has two flow stoppers 148, and the two flow stoppers 148 are arranged along the first direction D1.

[0143] The second semiconductor element 140 has a first bonding agent layer region 149A in which only the first bonding agent layer 126 is formed before the first semiconductor element 110 and the second semiconductor element 140 are bonded, and a second bonding agent layer region 150A which is a region where the second bonding agent layer 127 is formed on the first bonding agent layer 126. For example, the second semiconductor element 140 has one first bonding agent layer region 149A and two second bonding agent layer regions 150A. The first bonding agent layer region 149A extends along the second direction D2 between the first small bonding surface 146b and the second small bonding surface 146c. The second bonding agent layer regions 150A are respectively formed at the ends of the first bonding agent layer region 149A in the second direction D2 and at the ends of the first bonding agent layer region 149A in the direction opposite to the second direction D2. Flow stoppers 148 are respectively formed in the first direction D1 and the direction opposite to the first direction D1 in a part of the first bonding agent layer region 149A.

[0144] Figure 35 It shows a cross section in the case where the first semiconductor element 110 and the second semiconductor element 140 are cut along a plane extending in the first direction D1 and the third direction D3, showing the states before and after the first bonding agent layer 126 and the second bonding agent layer 127 are melted. As Figure 32 and Figure 35 shown, on each of the second direction D2 and the direction opposite to the second direction D2 of the first bonding agent layer 126, the molten second bonding agent layer 127 flows toward the center in the second direction D2 by heating.

[0145] In a state where the first bonding agent layer 126 and the second bonding agent layer 127 are melted, the opto-semiconductor device 1E includes a bonding agent layer 160 that bonds the third bonding surface 161 and the fourth bonding surface 162 to each other. The third bonding surface 161 is, for example, a part of the aforementioned electrode 17. More specifically, the third bonding surface 161 is a middle part of the electrode 17 in the first direction D1, and is a part of the electrode 17 that faces the first bonding agent layer 126 (the first bonding agent layer region 149A) along the third direction D3. The fourth bonding surface 162 is, for example, the electrode 28 of the second semiconductor element 140.

[0146] The bonding agent layer 160 is the first bonding agent layer 126 and the second bonding agent layer 127 that are melted and mixed. The volume of the bonding agent layer 160 is the sum of the volume of the first bonding agent layer region 149A and the volume of the second bonding agent layer region 150A. The area of the bonding agent layer 160 when observed along the third direction D3 is larger than the area of the joint when observed along the third direction D3. The area of the bonding agent layer 160 when observed along the third direction D3 is, for example, equal to or less than the area of the fourth bonding surface 162 when observed along the third direction D3.

[0147] For example, the area of the third bonding surface 161 when observed along the third direction D3 is equal to the area of the fourth bonding surface 162 when observed along the third direction D3. The area ratio of the bonding agent layer 160 when observed along the third direction D3 to the area of the third bonding surface 161 when observed along the third direction D3 is, for example, 1.1 or more and 5 or less.

[0148] As described above, in the opto-semiconductor device 1E according to the seventh embodiment, the second bonding surface 146 has a first small bonding surface 146b and a second small bonding surface 146c that are separated from each other along the first direction D1, which is a direction parallel to the second bonding surface 146. The first semiconductor element 110 has a third bonding surface 161 that is parallel to the first bonding surface 18 and extends along the second direction D2, which is a direction intersecting the first direction D1, between the first small bonding surface 146b and the second small bonding surface 146c. The second semiconductor element 140 has a fourth bonding surface 162 that faces the third bonding surface 161 and extends along the first direction D1. The opto-semiconductor device 1E further includes a bonding agent layer 160 that bonds the third bonding surface 161 and the fourth bonding surface 162 to each other. In this case, the third bonding surface 161 can be firmly bonded to the fourth bonding surface 162, so that the positional deviation when bonding the electrode 17 of the first semiconductor element 110 and the electrode 28 of the second semiconductor element 140 to each other can be reduced.

[0149] In the optical semiconductor device 1E, similar to the aforementioned optical semiconductor device 1D, for example, the third joint surface 161 and the fourth joint surface 162 are joined by a bonding agent layer 160 as solder. In the optical semiconductor device 1E, the third joint surface 161 is a part of the electrode 17, and a part of the electrode 17 is joined to the electrode 28 by solder. Therefore, compared with the case where the entire electrode 17 is joined to the electrode 28, the joining force of the third joint surface 161 with respect to the fourth joint surface 162 can be reduced, and thus the influence of solder joining on the hydrophilic joining can be reduced. In addition, by joining the third joint surface 161, which is the middle part of the first direction D1 of the electrode 17, to the fourth joint surface 162, it is possible to prevent the balance of the solder from being broken and the first semiconductor element 110 from tilting.

[0150] As described above, various embodiments and various modification examples of the optical semiconductor device and the manufacturing method of the optical semiconductor device according to the present invention have been described. However, the present invention is not limited to the aforementioned embodiments or modification examples, and can be further modified within the scope of the gist described in the claims. That is, the constitution, shape, size, material, number, and arrangement mode of each part of the optical semiconductor device according to the present invention, and the content and order of the processes of the manufacturing method of the optical semiconductor device can be appropriately changed within the scope of the aforementioned gist.

[0151] For example, in the aforementioned embodiment, an example in which the hydrophilic treatment is performed on both the first joint surface 18 and the second joint surface 26 has been described. However, the hydrophilic treatment of the second joint surface 26 can also be omitted. In the aforementioned embodiment and modification examples, various examples of the first joint surface and the second joint surface have been described. The material of the first joint surface may be indium phosphide (InP), indium gallium arsenide (InGaAs), gallium indium arsenide phosphide (GaInAsP), aluminum gallium indium arsenide (AlGaInAs), silicon dioxide (SiO2), or silicon nitride (SiN). The material of the second joint surface may be silicon (Si), silicon dioxide (SiO2), or silicon nitride (SiN). Thus, the material of the first joint surface and the material of the second joint surface can be appropriately changed.

[0152] In the above, various optoelectronic semiconductor devices according to the first to seventh embodiments, and the first to fifth modification examples have been described. The optoelectronic semiconductor device according to the present disclosure may also be composed of elements selected from the aforementioned first to seventh embodiments, and the first to fifth modification examples. In addition, the combination of the first semiconductor element and the second semiconductor element in the optoelectronic semiconductor device can be appropriately changed. For example, in the optoelectronic semiconductor device, any one of the first semiconductor elements 10, 10A, 10B, 10C, 40, 40A, 60, 80, 110 may be bonded to any one of the second semiconductor elements 120 and 140. That is, the combination of the first semiconductor elements 10, 10A, 10B, 10C, 40, 40A, 60, 80, 110 with respect to the second semiconductor elements 20, 50, 70, 90, 120, 140 can be appropriately changed.

Claims

1. An optical semiconductor device, wherein: The optical semiconductor device comprises: A first semiconductor element having a first bonding surface and an end surface intersecting the first bonding surface and capable of emitting an optical signal; and a second semiconductor element having a second bonding surface opposite to the first bonding surface and an optical waveguide extending in a direction parallel to the second bonding surface and capable of transmitting the optical signal; The first bonding surface and the second bonding surface are hydrophilically bonded to each other, The end surface of the first semiconductor element and the optical waveguide of the second semiconductor element are optically coupled to each other.

2. The optical semiconductor device according to claim 1, wherein An oxide film is formed at a joining portion where the first joining surface and the second joining surface are hydrophilically joined to each other.

3. The optical semiconductor device according to claim 1 or 2, wherein: The first semiconductor element has a third bonding surface parallel to the first bonding surface, The second semiconductor element has a fourth bonding surface opposite to the third bonding surface, The optical semiconductor device further includes a bonding agent layer bonding the third bonding surface and the fourth bonding surface to each other at a bonding portion. The area of ​​the bonding agent layer when viewed along a direction intersecting the third bonding surface, that is, a cross direction, is larger than the area of ​​the bonding portion when viewed along the cross direction.

4. The optical semiconductor device according to claim 1 or 2, wherein: The second joint surface has a first small joint surface and a second small joint surface separated from each other along a first direction parallel to the second joint surface, The first semiconductor element has a third bonding surface extending in a second direction parallel to the first bonding surface and intersecting the first direction between the first small bonding surface and the second small bonding surface. The second semiconductor element has a fourth joint surface that is opposite to the third joint surface and extends in the first direction, The optical semiconductor device further includes a bonding agent layer that bonds the third bonding surface and the fourth bonding surface to each other at a bonding portion.

5. A method for manufacturing an optical semiconductor device, wherein: The method for manufacturing an optical semiconductor device comprises the following steps: hydrophilizing the first bonding surface of the first semiconductor element; Aligning the second semiconductor element so that the second bonding surface and the first bonding surface are opposed to each other in a state of being separated from each other, and the end surface of the first semiconductor element capable of emitting an optical signal and the optical waveguide of the second semiconductor element capable of transmitting the optical signal are optically coupled to each other; bringing the first bonding surface into contact with the second bonding surface at a first temperature to press at least one of the first semiconductor element and the second semiconductor element to temporarily bond them to each other; as well as After the step of performing the provisional bonding, the first semiconductor element and the second semiconductor element are heated to formally bond the first bonding surface and the second bonding surface to each other at a second temperature higher than the first temperature.

6. The method for manufacturing an optical semiconductor device according to claim 5, wherein: In the step of performing the hydrophilization, the first bonding surface is hydrophilized by irradiating the first bonding surface with ultraviolet rays in an atmosphere-exposed environment.

7. The method for manufacturing an optical semiconductor device according to claim 5, wherein: In the step of performing the hydrophilization, the first bonding surface is hydrophilized by exposing the first bonding surface to oxygen plasma in a vacuum environment.

8. The method for manufacturing an optical semiconductor device according to claim 5, wherein: In the step of performing the hydrophilization, the first bonding surface is hydrophilized by exposing the first bonding surface to nitrogen plasma in a vacuum environment.

9. The method for manufacturing an optical semiconductor device according to any one of claims 5 to 8, wherein: The strength of a joining portion where the first joining surface and the second joining surface are joined to each other after the step of performing the provisional joining is 5 MPa or more.

10. The method for manufacturing an optical semiconductor device according to any one of claims 5 to 8, wherein: The distance between the first bonding surface and the second bonding surface during the alignment step is greater than or equal to 1 μm and less than 100 μm.

11. The method for manufacturing an optical semiconductor device according to any one of claims 5 to 8, wherein: The first temperature is 20°C or higher and 40°C or lower. The second temperature is 100° C. or higher and 300° C. or lower.

12. The method for manufacturing an optical semiconductor device according to any one of claims 5 to 8, wherein: The first semiconductor element has a third bonding surface parallel to the first bonding surface, The second semiconductor element has a fourth bonding surface opposite to the third bonding surface, Before the alignment step, the method includes forming a first bonding layer composed of an electrical bonding agent on the fourth bonding surface and forming a second bonding layer composed of the electrical bonding agent on the first bonding layer. The method includes the step of melting the first bonding layer and the second bonding layer by heating, and bonding the third bonding surface and the fourth bonding surface to each other via the melted first bonding layer and the second bonding layer, that is, the bonding layer.

Images